3 ms·
The post is about chasing the long tail of latency risk. A high -- even very high -- probability of not acquiring a lock is not good enough. Investigating libma
by jjoonathan 7y ago
The post is about chasing the long tail of latency risk. A high -- even very high -- probability of not acquiring a lock is not good enough. Investigating libmalloc and discovering that somehow it had strictly bounded, suitable execution time wouldn't be good enough either, because it could plausibly change without notice.
- vnorilo 7y agoTo get an idea of the length of the tail: Suppose a deadline of 6ms to produce an audio buffer. If 0.01% of the callbacks choke on a lock, we get a dropout once a minute, which is quite a lot. This means we're interested in the 99.99th percentile.